{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ZHROJZTEX5TCA3HZDDGQIDT7QE","short_pith_number":"pith:ZHROJZTE","schema_version":"1.0","canonical_sha256":"c9e2e4e664bf66206cf918cd040e7f812cb9a8fd5230353513f18c639b6c63bd","source":{"kind":"arxiv","id":"2111.01272","version":2},"attestation_state":"computed","paper":{"title":"Sequence Transduction with Graph-based Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Jonathan Le Roux, Niko Moritz, Shinji Watanabe, Takaaki Hori","submitted_at":"2021-11-01T21:51:42Z","abstract_excerpt":"The recurrent neural network transducer (RNN-T) objective plays a major role in building today's best automatic speech recognition (ASR) systems for production. Similarly to the connectionist temporal classification (CTC) objective, the RNN-T loss uses specific rules that define how a set of alignments is generated to form a lattice for the full-sum training. However, it is yet largely unknown if these rules are optimal and do lead to the best possible ASR results. In this work, we present a new transducer objective function that generalizes the RNN-T loss to accept a graph representation of t"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2111.01272","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-01T21:51:42Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"325b3f7d1a8b6f1eac009f2fbb907101a343912bf1cde6141dc0c47878cfefbd","abstract_canon_sha256":"07a5f6c00aa3b8e32987c0ca7e4143b6c476c679fc86bcaa0fdc7fd097dfdf1a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:10:24.438389Z","signature_b64":"qs9K365wmWzPBt8WfwCMP3z3+JPXcvl5u9vTUdLnLWedbqRP6AvzMkkM0+YBcwDcHIEynLpfckM5ug9x3ri/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9e2e4e664bf66206cf918cd040e7f812cb9a8fd5230353513f18c639b6c63bd","last_reissued_at":"2026-07-05T04:10:24.438036Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:10:24.438036Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sequence Transduction with Graph-based Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Jonathan Le Roux, Niko Moritz, Shinji Watanabe, Takaaki Hori","submitted_at":"2021-11-01T21:51:42Z","abstract_excerpt":"The recurrent neural network transducer (RNN-T) objective plays a major role in building today's best automatic speech recognition (ASR) systems for production. Similarly to the connectionist temporal classification (CTC) objective, the RNN-T loss uses specific rules that define how a set of alignments is generated to form a lattice for the full-sum training. However, it is yet largely unknown if these rules are optimal and do lead to the best possible ASR results. In this work, we present a new transducer objective function that generalizes the RNN-T loss to accept a graph representation of t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.01272","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2111.01272/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2111.01272","created_at":"2026-07-05T04:10:24.438091+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.01272v2","created_at":"2026-07-05T04:10:24.438091+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.01272","created_at":"2026-07-05T04:10:24.438091+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZHROJZTEX5TC","created_at":"2026-07-05T04:10:24.438091+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZHROJZTEX5TCA3HZ","created_at":"2026-07-05T04:10:24.438091+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZHROJZTE","created_at":"2026-07-05T04:10:24.438091+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE","json":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE.json","graph_json":"https://pith.science/api/pith-number/ZHROJZTEX5TCA3HZDDGQIDT7QE/graph.json","events_json":"https://pith.science/api/pith-number/ZHROJZTEX5TCA3HZDDGQIDT7QE/events.json","paper":"https://pith.science/paper/ZHROJZTE"},"agent_actions":{"view_html":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE","download_json":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE.json","view_paper":"https://pith.science/paper/ZHROJZTE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.01272&json=true","fetch_graph":"https://pith.science/api/pith-number/ZHROJZTEX5TCA3HZDDGQIDT7QE/graph.json","fetch_events":"https://pith.science/api/pith-number/ZHROJZTEX5TCA3HZDDGQIDT7QE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE/action/storage_attestation","attest_author":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE/action/author_attestation","sign_citation":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE/action/citation_signature","submit_replication":"https://pith.science/pith/ZHROJZTEX5TCA3HZDDGQIDT7QE/action/replication_record"}},"created_at":"2026-07-05T04:10:24.438091+00:00","updated_at":"2026-07-05T04:10:24.438091+00:00"}